<a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/pettankoutei"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4781d887-3f34-42b2-87b9-36c2c970d165/width=525/4781d887-3f34-42b2-87b9-36c2c970d165.jpeg" /></a>If you like my content consider supporting me! 😊1 Cookie = Early access to future releases! 🍪❤️<h3 id="heading-1124">v2.0 Update</h3>Compared to version 1.0, most of the information still applies. This is a NAI based model carefully merged to achieve an anime CG look. The style is heavily influenced by the negative prompt. The 'default' recommended prompt <code>(worst quality, low quality:1.4)</code> results in more realistic lighting and textures, but reduce the weight to 1.0 as in <code>worst quality, low quality</code>and it will lean more towards a '2.5D anime model'. See comparison <a target="_blank" rel="ugc" href="https://cdn.buymeacoffee.com/uploads/project_updates/2024/04/b8a0d496b459f5ab4c34e0cdebca7a65.png">here</a>.Other ways you can influence the style is by using quality tags in the positive prompt to push its style a little bit either to anime/illustration or to semi-realism with realistic lighting and textures. When using 2D/anime character LoRAs an useful example would be tags such as (RAW photo, realistic, photograph, shot with Fujifilm X-Pro), which despite what they may imply, it just helps produce more realistic-looking textures, hair-strands, etc, when the 2D style of the LoRA is too heavy. It still is very much an anime inspired model, intended for CG-like images.This will be the last update to this series at least for the time being: I'm happy with them for now and I'll be experimenting with other models, while working on more style loras for Pony and SD1.5. Follow me on <a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/pettankoutei">my BMaC page</a> for updates and consider supporting me if you liked my models: any donation is elegible to early access and some exclusive content.<hr /><h2 id="heading-4">About</h2>This series is designed for colorful and cute characters. It is very controllable and responds very well to your prompts. LoRA compatibility is good enough in my opinion, but as always it is dependant on the LoRA in question and how well you use them.Recommended parameters:Positive prompt: (masterpiece, best quality, SFW:1.2)Negative prompt: (worst quality, low quality, NSFW:1.4), nude, topless, bottomlessClip skip: 2Sampling method: Euler a, DPM++ 2M Karras, DPM++ SDE Karras (suggested, use whatever you prefer)Sampling steps: 20 (minimum)CFG Scale: 7 (default during tests)Hires. fix:Upscaler: 4x-AnimeSharp (recommended)Hires steps: 10 (minimum)Denoise: 0,34Upscale by: 1.3 (minimum), 2 (suggested)In case you are not using hires. fix, or if characters are far away, ADetailer (facedetailer) is always recommended.DisclamerAll the models that I have developed and published are for entertainment purposes only. They are derived from the base Stable Diffusion 1.5 model, which means it also includes its inherent issues, limitations and unfiltered capabilities. With that in mind, I am in no way responsible for any content that the user creates with malicious intent while using any of these models. The user assumes complete responsibility for the misuse of these tools and any unethical content created with them. This disclaimer is subject to the laws of the United States of America and the state of California. I reserve the right to remove or report any content that violates the terms of service, ethical standards, or applicable laws.License conditionsThe model published here can be used for personal, non-commercial purposes only. It includes several publicly available resources in its development, each with various types of licenses. It is discouraged to upload this model for use in online generation services: I am not responsible for any license infringements.

Pettanko RouraMashin SR

Version 2.0 includes recent changes made to the base 2D v2.0 model:<ul><li>Included recent updates to the models used in the base merge.</li><li>Included newest changes made to my personal fine tune.</li><li>Minor readjustments to the block merging process.</li></ul>Compared to v1.0, this new version presents both advantages and some disadvantages. It is easier to generate more dynamic and aesthetic camera angles, but hands for example are a little less stable. Additional changes were made during fine tuning merge to help stabilize hands and general anatomy and the result seemed satisfactory.

Files with SHA256: 1b3e344a881eb05c12b7ba48b81573b939e7079b728fc69195aa7d5c79ab65f3